# `Monty.Frameworks.Models.EvidenceMatching.FeaturesForMatching.Selector.DefaultFeaturesForMatchingSelector`
[🔗](https://github.com/d-led/ex.tbp.monty/blob/v0.1.0/packages/ex_tbp_monty_core/lib/monty/frameworks/models/evidence_matching/features_for_matching/selector/default_features_for_matching_selector.ex#L1)

Features are matched when the experiment asked for them.

Mirrors
`tbp/monty/frameworks/models/evidence_matching/features_for_matching/selector.py::DefaultFeaturesForMatchingSelector`.

Two conditions, both of which have to hold for a channel to be matched on its
features: the increment that scales feature evidence is above zero, and the
channel weights at least one feature that is not a pose feature. A channel
weighted only on `pose_vectors` is a channel matched by pose alone — that is
a real way to run an experiment, not an omission.

---

*Consult [api-reference.md](api-reference.md) for complete listing*